Tools 81

Gives bots a unified way to read public web pages, RSS feeds, and authorized media, with channel diagnostics.

Lets bots extract media metadata, subtitles, audio, or video from supported sites into the workspace.

A personal trading agent for market research, analysis, and backtesting, with 54 MCP tools registered for the bot.

Lets bots read and write Word, Excel, PowerPoint, and PDF files, with rendering, validation, and conversion.

Builds a structured repository index so bots can understand symbols, relationships, and code context across sessions.

Microsoft file-to-Markdown conversion for PDF, Office, images, audio, HTML, CSV, and archives.

Language-server-backed symbol retrieval, reference lookup, and semantic editing for coding agents.

A universal document converter for moving content between Markdown, HTML, DOCX, EPUB, and many other formats.

A personal tutor that decomposes questions, researches sources, and builds step-by-step explanations.

Turns outlines, data, and scene plans into 1080p H.264 MP4 videos.

Converts complex PDFs, scans, formulas, and tables into agent-friendly Markdown.

Agent-oriented web crawling and structured extraction with dynamic-page and Markdown support.

Transcribes OpenClaw or Hermes voice messages with CPU int8 and the lightweight base model.

Chinese and multilingual speech recognition with timestamp-capable output.

High-level browser task planning for websites, forms, and multi-step browsing workflows.

Parses PDF, Office, HTML, and images into structured documents, Markdown, or JSON.

Maintains searchable, updateable user memories across agents with a self-hosted server option.

Visual workflow automation with native AI steps, custom code, and hundreds of application connectors.

Connects agents to Linear issues, projects, teams, comments, and product workflows through the official hosted MCP server.

Aggregates results from multiple search services without profiling users and exposes a JSON search API.

Builds production agent applications with visual workflows, retrieval, model routing, and observability.

A self-hosted interface for local and hosted models with tools, retrieval, pipelines, and user management.

A broad framework for composing model calls, tools, retrieval, memory, and agent workflows.

Combines document understanding, retrieval, citations, and agent workflows for knowledge-heavy tasks.

Coordinates research, coding, content creation, sandboxes, memory, tools, and subagents for long-running work.

Creates secure, elastic development sandboxes for running AI-generated code and agent workloads.

Handles single requests through full crawls with adaptive selectors and anti-bot-aware browser fetching.

Provides document workspaces, retrieval, agents, tools, and multi-model chat in a self-hosted environment.

Coordinates agent swarms, autonomous workflows, shared memory, retrieval, and coding-agent integrations.

Supplies version-specific library documentation and code examples to coding agents through MCP.

Builds agent and retrieval workflows visually, then exposes them as APIs and embedded assistants.

Connects agents to files, databases, APIs, indexes, retrieval pipelines, and document extraction.

Microsoft framework for building event-driven, conversational, single-agent, and multi-agent applications.

Orchestrates role-based agents and deterministic flows for collaborative, multi-step business tasks.

Plans research, gathers web and local sources, synthesizes evidence, and produces cited reports.

Combines no-code automation, AI agents, MCP tools, and hundreds of integration pieces.

Automates Chromium, Firefox, and WebKit with reliable page actions, contexts, downloads, and screenshots.

Controls Chrome and Firefox through a high-level JavaScript API for navigation, extraction, and rendering.

A mature cross-browser automation ecosystem with WebDriver support across languages and grid execution.

Gives coding agents Chrome DevTools access for page inspection, network analysis, performance, and automation.

Connects agents to repositories, code, issues, pull requests, actions, releases, and GitHub security data.

Downloads, manages, and serves a broad catalog of local language and vision models through a simple API.

Runs quantized language models efficiently across CPU, Metal, CUDA, Vulkan, and other hardware backends.

Serves language models with high throughput, efficient memory use, batching, and OpenAI-compatible APIs.

Normalizes model providers behind OpenAI-compatible APIs with routing, budgets, guardrails, and logging.

A high-performance vector search engine with filtering, payloads, distributed deployment, and client libraries.

Scalable vector similarity search for large embedding collections, hybrid retrieval, and production RAG.

Stores embeddings, documents, and metadata behind a developer-friendly retrieval API for AI applications.

Combines vector, keyword, and structured filtering over stored objects in a cloud-native database.

Turns images and PDFs into structured text and document data across more than one hundred languages.

Provides a broad server-side PDF toolbox for conversion, OCR, redaction, signing, splitting, and merging.

Runs Whisper transcription efficiently across CPU, Metal, CUDA, Vulkan, and other local backends.

Captures traces, evaluations, prompts, datasets, metrics, and experiments for production AI systems.

Builds a persistent structural map of a repository so coding agents can retrieve focused review and impact context.

Routes OpenAI-compatible requests across model providers with quota-aware fallback, compression, and policy controls.

Turns Python functions and services into production MCP tools, resources, prompts, clients, and interactive apps.

Builds persistent agent memory from documents and interactions using graph, vector, and semantic retrieval.

Runs large mixture-of-experts models with CPU-GPU heterogeneous inference and an OpenAI-compatible server.

Runs a coding agent through a headless OpenAPI server with sessions, messages, files, providers, and events.

Builds and deploys agent workflows visually, then exposes them through authenticated run, webhook, and OpenAI-compatible APIs.

Runs modular node workflows for image, video, and audio generation through a queue-based server API.

Runs software agents on isolated local or remote workspaces through a production REST and WebSocket agent server.

Incrementally builds time-aware knowledge graphs for agents, with hybrid retrieval over evolving facts and relationships.

Adds local speech generation, transcription, cloned voice profiles, and agent voice output through REST and MCP.

Generates speech in 31 languages with a lightweight 99M-parameter ONNX model through native and OpenAI-compatible HTTP endpoints.

Gives agents keyword, SERP, backlink, rank-tracking, and Search Console research through a hosted or self-hosted MCP server.

Combines product analytics, session replay, feature flags, experiments, data pipelines, and an agent-ready MCP interface.

Routes Claude, OpenAI, Gemini, and Grok accounts through a unified, multi-user API with quotas and scheduling.

Coordinates issues, projects, workspaces, autonomous runs, and coding-agent runtimes through a REST and WebSocket backend.

Lets agents operate a local-first Chromium browser with existing sessions through more than 50 built-in MCP browser tools.

Provides a compact self-hosted backend with realtime collections, authentication, file storage, and a complete REST API.

Runs queued Google Maps lead and place extraction jobs through a documented REST API with CSV downloads.

Turns local hardware into a private AI stack with llama-server inference, chat, voice, RAG, automation, and operations tools.

Gives coding agents one gateway for provider routing, failover, model profiles, client quotas, tools, and request observability.

Gives agents live geopolitical, market, conflict, aviation, infrastructure, climate, and news intelligence through a hosted MCP server.

Keeps durable agent memories in a single SQLite file with hybrid vector, full-text, and importance ranking, shared across MCP clients.

Stores agent memories in an entity graph with hybrid semantic and keyword recall, then consolidates duplicates and stale facts in the background.

Builds a durable model of each user across sessions, exposing peer representations and cards the agent can query for preferences and context.

Connects agents to Jina AI’s official hosted MCP tools for reading URLs, web and image search, reranking, embeddings, and PDF extraction.

Reads and searches PDFs with OCR, table extraction, document structure, page citations, and focused visual evidence tools.

Searches, fetches, crawls, extracts, caches, and researches the web through an authenticated remote MCP endpoint.